Rapid Screening and Prioritization of Culture Conditions for Natural Product Discovery using the Liquid Microjunction Surface Sampling Probe

The discovery of novel bioactive compounds remains a cornerstone of natural product (NP) chemistry. However, a notable bottleneck in traditional NP discovery workflows is the extraction and purification of compounds of interest, a time and resource intensive process, which hinders sustainability and efficiency of novel hit discovery in multi-condition screening projects. This study evaluates the application of the liquid microjunction surface sampling probe (LMJ-SSP) with partial least squares discriminant analysis (PLS-DA) as a new step prior to the NP discovery workflow. By integrating ambient mass spectrometry with machine learning, we were able to analyze four strains of Penicillium fungi grown under thirteen unique conditions in situ, without sample preparation. By leveraging PLS-DA, the workflow prioritized the growth conditions that maximized chemical diversity, offering insights into metabolite composition before more resource intensive bulk growth and chromatographic methods are applied. The LMJ-SSP based approach also achieved a significant reduction in sampling time (96%), overall cost (98%), and solvent consumption (98%). Our triaging method efficiently streamlines the NP discovery pipeline through chemically-informed prioritization, reduced rediscovery of known compounds, and improved sustainability.